Boeing Hat Section Study

7/29/2019 Boeing Hat Section Study

1/17

Hy-Bor

for Advanced Hat-stiffened Aircraft Structures

A design and verification study was undertaken by the Boeing Phantom Works, St. Louis,MO to evaluate the potential structural benefits and cost impact of the selective use ofHy-Bor

hybrid boron-graphite prepreg tape in an advanced aircraft design.

Emerging composite hat stiffened skin designs for combat aircraft are using a designapproach where the inner moldline skin, web, and cap plies are continuous. This designapproach has shown a performance benefit over discrete flange concepts, especially inhigh compression load applications. The objective of this study was to determine thestructural efficiency benefits of Hy-Bor

in composite material design and to perform a

manufacturing cost analysis. Using loads typical of a highly-loaded compression regionof an advanced aircraft wing skin, this analysis task used Boeing software tools tooptimize a composite baseline hat section design.

Typical geometric parameters considered for optimization are shown in the Figure 1. The

baseline design consisted of Cytecs T40-800/5215 tape and T300 5HS/5215 fabric.Starting with the optimized baseline design, Hy-Bor

was integrated into the cap and

base regions to develop an optimized Hy-Bor reinforced hat design. It was expectedthat the Hy-Bor reinforced hat design would allow for an increase in hat spacing(Figure 2), a reduction in skin thickness, or reduction in hat size leading to significantweight savings. In addition to determining potential weight savings, a manufacturingcost comparison between the baseline and Hy-Bor reinforced concepts was conducted.

Cap Width

Base Width

Base Thickness

Web

Thickness

Cap Thickness

Height

Skin Thickness

Nugget Filler

Potential Hy-Bor Application

Web Angle

Figure 1 - Baseline & Hy-Bor Reinforced Design Concepts
http://specmaterials.com/


2/17

Baseline Skin Panel Hat Spacing

Hy-Bor Rinforced Skin Panel Hat Spacing

Figure 2 - Software Optimization at Panel Level

Based on initial design requirements, the baseline skin thickness was set at 0.13-inchthickness (min.) with a mandrel height of 2" and mandrel cap width of 1.5". Based onthese requirements and design loads, a baseline concept was determined.

Using this baseline concept, Hy-Bor

plies were then added to both the hat cap and theskin area under the hat. Five variations in skin laminates ranging from 0.1302" - 0.141"in thickness were considered. The addition of Hy-Bor plies in into the skin had noeffect on weight, mostly due to the closeness of the skin panel to the structures neutral

axis. A weight savings of only 1% (0.2 lbs for the 17.5 lb study panel section) wasrealized.

Thinner skin laminates, ranging from 0.1192 to 0.1248-inch, were then evaluated. WithHy-Bor plies only in the cap regions of the hats, the weight savings were still in the 1%range. The hat height was then varied between 1 and 2-inches at 0.25-inch increments.The hat cap width was varied in the same manner. The hat web angle remained the samesince this angle was found to be optimum based on studies conducted on previous Boeingprograms. Three skin laminates at two thicknesses (0.1194 and 0.1302-inch) wereconsidered. As before, Hy-Bor

plies were placed in the cap of the hat. At this point a

weight savings of 0.8 lbs or nearly 5% on a panel basis was realized. Final panelgeometries for the Baseline and Hy-Bor designs are shown in Figure 3.

Considering an advanced aircraft platform with approximately 3000 lbs integrated hat-stiffened composite structure, this translates into an appreciable 150 lbs of weightsavings. A potential secondary benefit would be the 28% smaller hat height that wasobtained. In a wing portion of such a design the smaller hat sections could allowincreased fuel capacity.


3/17

t (in) Layup t (in) Layup Length (in) Width (in) Stiffeners

0.119445

c, 0

c, +45, -45, 03, 90,

03, -45, +45, 0c, 45

c0.06 45

c, 0

c, 0

c, 45

c 40 45 6

Skin Web Panel

Common Features

t (in) Layup Reinforcement Base (in) Cap (in) Height (in)

Baseline 0.1032 45c, 03, 0

c, 02, 0

c, 03, 45

c 0 Tape 1.62 1 1.75 17.5

Hy-Bor 0.1149 45c, 0

H3, 0

c, 0

H3, 0

c, 0

H3, 45

c 0 Hy-Bor 1.44 1 1.25 16.7

Weight

(lbs)Cap Mandrel

Optimized Features

Configuration

Skin

Web Cap

Mandrel

Base

MandrelCap

MandrelHeight

45 Cloth

0 Cloth

Reinforcement

Tape Plies Skin

Web Cap

Mandrel

Base

MandrelCap

MandrelHeight

45 Cloth

0 Cloth

Reinforcement

Tape Plies

C = cloth H = Hy-Bor

Figure 3 Optimization Summary

A cost comparison was completed for the baseline hat stiffened panel design and the Hy-Bor

hat stiffened panel design. Design and manufacturing information for each design

was loaded into Boeings DFM software and labor outputs were compared. Inputs


4/17

consisted of the number of skin plies and skin lay-up information, and the number of hatsand hat ply lay-up information.

It was assumed that the same outer mold line (OML) tool would be used for both designsand would not impact the cost comparison; therefore, only mandrel tooling was

estimated. Estimates for the mandrel tooling were made by tool engineering rather thanusing the DFM software.

A summary of the cost comparison breakdown between the baseline and Hy-Bor

designs is presented in Figure 4. Assumptions used in the cost estimating process arelisted below:

1. The part lay-up was broken down as follows to get the most accurate estimate:

Skin -- Two plies of cloth and 11 plies of tape

Hat web -- Two plies of cloth wrapped around the mandrel

Hat cap -- 5215 tape or Hy-Bor tape depending on which design

Over lay plies -- 2 Cloth plies laid over the skin and hats

2. Cytec material cost per pound for small quantities supplied by Cytec:

T40-800/5215 tape -- $81.63/lb

T300 5HS/5215 cloth $49.08/lb

3. Calculations were made to determine a cost in $/lb for Hy-Bor prepreg with a 5215resin as follows:

A. Cost of T40-800/5215 tape was converted from $81.63/lb to $1.73/LF$81.63/lb x 0.055lb/in3 = $4.55/in3$4.55/in3 x 0.0054in (5215 tape thickness) x 144in2/ft2 x 0.5ft

(width of tape) = $1.73/LF

B. SMI cost of $17.75/LF and added to cost of 5215 tape of $1.73/LF5215 Hy-Bor = $19.48/LF

C. Cost of 5215 Hy-Bor was converted from $19.48/LF to $693/lb$19.48/LF x 1LF/12in length x 6in width x 0.0061in thickness =$19.48/LF x 1/LF/0.4392in3 x 1in3/0.064lb = $693/lb

4. There were no manufacturing differences between the two designs; the onlydifferences were the hat cap material and hat sizes. The DFM results show a slightcost savings for the Hy-Bor

design because of less 5215 material cost and slightly


5/17

less labor (due to smaller hat size) but a significant cost increase in the hat capmaterial as compared to the baseline hat cap material cost.

Baseline Design Hy-Bor DesignHy-Bor Cost

Increase

SKIN

Material Cost $631.00 $631.00 $0.00

Labor cost @ $150/hr $781.00 $781.00 $0.00

Total skin cost $1,412.00 $1,412.00 $0.00

HATS*

Material Cost $166.20 $680.40 $514.20

Labor Cost @ $150/hr $3,612.00 $3,726.00 $114.00

Total Hat cost $3,778.20 $4,406.40 $628.20

TOP 2 PLIES OVER

HATS AND SKIN**Material Cost $200.00 $185.00 -$15.00

Labor cost @ $150/hr $414.00 $395.00 -$19.00

Total Top 2 Plies Cost $614.00 $580.00 -$34.00

TOTAL MATERIAL COST $997.20 $1,496.40 $499.20

TOTAL LABOR COST $4,807.00 $4,902.00 $95.00

TOTAL COST $5,804.20 $6,398.40 $594.20

*Added cost of web plies and cap plies and multiplied by 6 hats

** These are the 2 cloth plies that go over the hats and continue over the skin. They were not

accounted for in the skin or hats.

Figure 4 - Cost Comparison Summary


6/17

The structure used as the basis of this trade study represented an advanced airframeconcept containing approximately 3,000 lbs. of hat-stiffened skin composite structure.Application of Hy-Bor to the skins of the structure did not result in a significant weightsavings; however, a weight saving of 5% (150 lbs. total) was achieved with the inclusion

of Hy-Bor

in the hat components and appropriate resizing. The reduction in hat depthresulting from increased compression properties in the cap region provides anopportunity for significant increase in fuel capacity in the wing portions of the structure.On a total baseline structural weight of 3,000 lbs., the projected materials andmanufacturing cost to save 150 lbs. would be $127,575, or $850.50/lb of weight saved.The value of increased internal wing volume for fuel was not estimated.

In the verification portion of this study a Hy-Bor reinforced, hat stiffened panel (Figure5) was fabricated and machined to produce three hat stiffened panel subcomponents forstructural testing in combined compression-pressure.

27.75

40.0

7.3

Panel width allowsfor subcomponentmachining

Subcomponent width is 6.5.0.4 added to eachside for simple-supportfixturing.

Figure 5 - Subcomponent Plan for Hat-Stiffened Panel


7/17

The panel fabrication consisted of a series of debulking operations to removeentrapped air and aid in compaction. Figure 6 shows one such operation that occurredafter the majority of the skin, hat mandrels, inner wrap hat plies, and first three Hy-Borcap plies were located.

Figure 6 - Hat-Stiffened Panel During Typical Debulking Operation

(Skin, Hat Mandrels, Inner Wrap Hat Plies and First Three Hy-Bor Cap Plies in Place)

After curing, the panel underwent a post cure at 350F for four hours. The hatends were trimmed and the panel has been sent to NDI prior to subcomponent extraction.The completed stiffened panel is show from the inner mold line and outer mold line inFigures 7 and 8, respectively.


8/17

Figure 7 - Inner Mold Line of Panel with Trimmed Hats Prior to NDI

Figure 8 - Outer Mold Line of Panel with Trimmed Hats Prior to NDI

Ultrasonic nondestructive inspection revealed that no significant indications werepresent and the panel was of production quality. The test subcomponents were thenmachined from the panel (Figure 9) and configured with load adapters in preparation forinstrumentation and test (Figure 10).


9/17

Figure 9 - Machined Subcomponents


10/17

Figure 10 - First Subcomponent Ready for Test

(Load Adapters, Tension Pads, and Instrumentation in Place)

The compression test set-up is shown in Figure 11 with a close up of thesubcomponent from the inner moldline (IML), stiffener side, shown in Figure 12. Thesimulated air pressure of 5 psi was applied to the tension pads, Figure 13, through a dead-weight setup which simply utilized a mobile hydraulic lift which held the dead weight inplace and slowing released the weight to apply load to the tension pads (Figure 14).

Figure 11 - Subcomponent Test Setup


11/17

Figure 12 - Close-up of Subcomponent 1 during Test

IML (Stiffener Side)


12/17

Figure 13 - Tension Pads on OML Surface of Subcomponent 1


13/17

Figure 14 - Simulated Pressure Applied Through Dead Weight Setup

Design ultimate compression load (DUL) for the subcomponent was 16, 500 lbs.Subcomponent 1 achieved that level with no audible indications of failure events.Continuation of the test resulted in failure at 22, 555 lbs of compression load with no

audible indications of failure until the failure load was reached; audible cracking andassociated response in strain data indicated failure had occurred. Failure occurred in thehat run-out region at the top of the subcomponent, Figure 15. This failure is typical ofthis advanced hat design concept which precludes failure from occurring at the hatflange/skin location at lower loads since there is no discrete hat flange/skin interface.


14/17

Figure 15 - Subcomponent 1 Failure Hat Run-Out Region

The compression with pressure load testing was subsequently conducted onsubcomponents 2 and 3. The failing loads for all three subcomponents were:

Subcomponent 1 22,560 pounds Subcomponent 2 16,180 pounds Subcomponent 3 32,080 pounds

While this is a considerable spread in the test results, the failure mode, delamination inthe hat taper run-outs, was similar for all subcomponents. Failure locations for each of thesubcomponents relative to key strain gage locations are shown in Figure 16. Axial strain gageresponse on each side of the hat run-out at each end of the subcomponents is shown in Figures 17through 19 for Subcomponents 1, 2, and 3, respectively. The deviation in strain at loading onsetis due to the applied 5 psi pressure load.

CLCL

8A8A

1717 11A11A1818

14A14A 7A7A

12A12A

13A13A

Subcomponent 1

Hat Taper Run-out Failure

Location

Subcomponents 2 & 3

Hat Taper Run-out Failure

Location

Odd Numbered Strain Gages on Hat Side

Figure 16 - Subcomponent Failure Location


15/17

Subcomponent 1

-24000

-22000

-20000

-18000

-16000

-14000

-12000

-10000

-8000

-6000

-4000

-2000

0

-3500-3000-2500-2000-1500-1000-5000500100015002000

Strain in/in

Load

lbs

Gage 12A PAD SIDE

Gage 17 HAT SIDE

Gage 7A HAT SIDE

Gage 11A HAT SIDE

Gage 18 PAD SIDE

Gage 8A PAD SIDE

Gage 13A HAT SIDE

Gage 14A PAD SIDE

12A12A

8A8A1717

7A7A

11A11A

1818

14A14A

13A13A

Figure 17 - Axial Strain Gage Results Subcomponent 1

Subcomponent 2

-24000

-22000

-20000

-18000

-16000

-14000

-12000

-10000

-8000

-6000

-4000

-2000

0

-3000-2500-2000-1500-1000-500050010001500

Strain in/in

Load

lbsGage 12A PAD SIDE

Gage 17 HAT SIDE

Gage 7A HAT SIDE

Gage 11A HAT SIDE

Gage 18 PAD SIDE

Gage 8A PAD SIDE

Gage 13A HAT SIDE

Gage 14A PAD SIDE

8A8A 171711A11A181814A14A

7A7A

12A12A

13A13A



16/17

Subcomponent 3

-36000

-32000

-28000

-24000

-20000

-16000

-12000

-8000

-4000

0-3500-3000-2500-2000-1500-1000-50005001000

Strain in/in

Load

lbs

Gage 12A PAD SIDE

Gage 17 HAT SIDE

Gage 7A HAT SIDE

Gage 11A HAT SIDE

Gage 18 PAD SIDE

Gage 8A PAD SIDE

Gage 13A HAT SIDE

Gage 14A PAD SIDE

8A8A

1717

11A11A

1818

14A14A

7A7A

12A12A

13A13A


The strain response indicates the presence of beam-column action which is expected forcombined compression-pressure loading. The results indicate that the degree of bending straincaused by beam-column action has a profound effect on failure load. It should also be noted thatthe strain results indicate that subcomponents stiffness were similar.

Prior to this project Boeing conducted an in-house program to develop an advanced allcarbon/epoxy concept that was designed to similar loading and tested in compression with 5 psisimulated pressure. Although all the details of this design cannot be disclosed in this paper,enough detail can be shown to make a strength vs. unit volume weight comparison, Figure 20. Itshould be noted that the highest (32,080 lb.) Hy-Bor test result was used in the comparison.Also, the advanced hat concept value is based on one test.

Key information that can be observed from this comparison is:

Strengths per Unit Volume Weight are very similar ~ 500 Kips/lb.

Hy-Bor design is over 8% lighter. Reduced hat height will result in increased fuel or payload capacity.


17/17

Hy-Bor Hat

Advanced Concept Hat

7.20

2.10

7.3

1.48

Area = 1.144 in2

Area = 1.255 sq in

C/E = 0.055lb/in3 Hy-Bor = 0.064lb/in3

Unit Volume Weight

= 0.0634 lbs

Unit Volume Weight

= 0.0690 lbs

Carbon/Epoxy = 1.089 in2

Hy-Bor = 0.055 in2

Highest Failing Load

= 32.08 kips

Highest Failing Load

= 34.60 kips

Strength/UVW= 506 kips /lb

Strength/UVW

= 501 kips /lb

Figure 20 - Strength per Unit Volume Comparison

Conclusion

This design and verification study was conducted to demonstrate the structural and costeffectiveness of hybrid boron/graphite composite tape (Hy-Bor) in an advanced composite hatstiffened skin aircraft structure. The design study revealed that, for the specific structure in thestudy, Hy-Bor could be used in the cap portion of the integral hat stiffeners to reduce weight by5%, at an additional cost of $851 per pound of weight saved. An additional benefit to the studydesign was a 28% reduction in hat height that might translate into a usable volume increase. Theverification portion of the study confirmed the ability to meet and exceed the DUL. Comparisonto a baseline design with equal strength per unit volume capability reveals an 8% weight savings.

Documents

Boeing Hat Section Study